Re: merge --no-ff is NOT mentioned in help

From: Jeff King <peff@peff.net>
Date: Nov 17, 2016, 22:21 UTC
Message-ID: <20161117222142.mca6lmhj5mvl4gbp@sigill.intra.peff.net>
In-Reply-To: <xmqqr36anibl.fsf@gitster.mtv.corp.google.com>

On Thu, Nov 17, 2016 at 09:10:22AM -0800, Junio C Hamano wrote:

> People interested may want to try the attached single-liner patch to
> see how the output from _ALL_ commands that use parse-options API
> looks when given "-h".  It could be that the result may not be too
> bad.

The output is less ugly than I expected, but still a bit cluttered IMHO. I was surprised that the column-adjustment did not need tweaked, but the code correctly increments "pos" from the return value of fprintf, which just works.

Looking at the output for --ff, though:

   --[no-]ff             allow fast-forward (default)

I do not think it's improving the situation nearly as much as if we made the primary option "--no-ff" with a NONEG flga, and then added back in a HIDDEN "--ff". I thought we had done that in other cases, but I can't seem to find any. But it would make "--no-ff" the primary form, which makes sense, as "--ff" is already the default.

Another option would be to teach parse-options to somehow treat the negated form as primary in the help text. That's a bit more code, but might be usable in other places.

-Peff